Test for zombie, memory, and pipe leaks

[richardsheridan]

I had originally assumed that multiprocessing was going to take care of all this for us, but in #34 @zgoda claims that hypercorn --debug -k trio somehow leads to a semaphore leak on each run_sync invocation. Since we are doing Readme Driven Development and claim "no leaks" we should develop the automated tests to show that there are no leaks in a suite of example usages and regression tests.

[richardsheridan]

We removed all semaphores in #33 so at least that's sorted.

[richardsheridan]

There is at least one situation where we have a pipe leak, which is whenever an ACK is not received on windows:

trio-parallel/trio_parallel/_proc.py

Lines 72 to 93 in 22b96bd

	if sys.platform != "win32":
	return
	# Give a nice error on accidental recursive spawn instead of hanging
	async def wait_for_ack():
	try:
	code = await self._receive_chan.receive()
	assert code == tp_workers.ACK
	except BaseException:
	self.kill()
	with trio.CancelScope(shield=True):
	await self.wait() # noqa: TRIO102
	raise
	nursery.cancel_scope.cancel()
	exitcode = None
	async with trio.open_nursery() as nursery:
	nursery.start_soon(wait_for_ack)
	exitcode = await self.wait()
	nursery.cancel_scope.cancel()
	if exitcode is not None:
	raise BrokenWorkerProcessError("Worker failed to start", self.proc)

I consider this an upstream bug (python/cpython#114720), but basically I would have caught this back in 2021 with these kind of tests.